Tuberculous (TB) meningitis is a serious, life-threatening disease affecting vulnerable populations including HIV-infected individuals and young children. Early diagnosis is challenging and outcomes are poor even with prolonged antimicrobial treatment (?12 months). Although several key antimicrobials have limited central nervous system (CNS) penetration and immunopathology is the major driver of neurological damage, pulmonary TB is still used as the treatment paradigm, with limited efforts to utilize preclinical models of TB meningitis to optimize treatment. We have developed a rabbit model of TB meningitis that replicates key neuropathological features of human disease. Subarachnoid infection with M. tuberculosis led to characteristic exudative meningitis and granulomatous brain lesions along with delayed maturation and neurologic deficits in the juvenile rabbits, similar to those seen in children with TB meningitis. Positron emission tomography (PET) with 124I-DPA-713, a novel tracer to noninvasively measure microglial inflammation, demonstrated intense uptake in infected-brain lesions. Recently, we have also evaluated daily multi-drug (rifampin and isoniazid) treatment in this model. Given the importance of rifampin to establish cure, better efficacy with high-dose rifampin in patients with TB meningitis and the limited data on antimicrobial concentrations in brain lesions (where bacteria reside), we also utilized dynamic 11C-rifampin PET in live rabbits and post-mortem analyses to determine rifampin exposures. Rifampin penetration into infected-brain lesions was limited, spatially heterogeneous and decreased substantially (32% to 11%) within two weeks of starting treatment. Importantly, antimicrobial concentrations in cerebrospinal fluid (CSF) did not correlate well with those in infected-brain lesions. First-in-human 11C-rifampin PET in a TB meningitis patient also demonstrated similar limited and heterogeneous rifampin penetration. Our overarching hypothesis is that multiple different pathological states occur simultaneously in the same subject, each with distinct bacterial burden, immunological state, and antimicrobial exposure, which also change with disease progression and treatment. Therefore, understanding the local pathology in infected-brain lesions is critical for developing effective treatments and limit neurological disabilities. We will utilize the rabbit model of TB meningitis and in vivo imaging to: a) measure the spatial distribution of new TB drugs in infected- brain lesions; b) compare linezolid and high-dose rifampin based regimens with standard TB treatment and; c) perform multi-modality imaging to simultaneously visualize intralesional bacterial burden, inflammation and antimicrobial exposure during TB treatments in live animals to correlate the effect of intracerebral inflammatory responses and antimicrobial exposures with the treatment outcome in the same animals. These assessments are not feasible with current technologies that require resected tissues. This proposal fulfills an important gap in TB drug development and treatment optimization for a devastating disease affecting vulnerable populations.